WO2009026642A1

WO2009026642A1 - System and method for precise real-time measurement of a target position and orientation relative to a base position, and control thereof

Info

Publication number: WO2009026642A1
Application number: PCT/AU2008/001275
Authority: WO
Inventors: Mark Joseph Pivac; Michael Barrington Wood
Original assignee: Goldwing Nominees Pty Ltd
Priority date: 2007-08-28
Filing date: 2008-08-28
Publication date: 2009-03-05
Also published as: AU2008291702A1; WO2009026641A1; CA2732312A1; EP2244865A1; EP2244865A4; CA2732310A1; AU2008291701A1; EP2249997A1; EP2249997A4

Abstract

An apparatus for precise real-time measurement or measurement and control of the position and orientation of a target (105) relative to the position of a base (102) is disclosed. The base (102) has a distance sensor (101) with angle encoders arranged to send a beam to the target (105) to measure distance, altitude, and azimuth and to output spatial position data of the target. The target (105) has a target orientation sensor (103) mounted on a pan and tilt mechanism (113, 115), and is arranged through the pan and tilt mechanism (113, 115) to reflect the beam back to the distance sensor (101). The target orientation sensor (103) outputs orientation data being a measurement of orientation of the target (105). The apparatus also has an inertial reference system (104) including motion sensors (125, 127, 129, 131, 133, 135) in or on the target (105) to measure linear and angular acceleration in the target (105). Data is combined in processor circuitry (100) to derive position and orientation of the target (105) with respect to time, which can be used as input in the control of the position and orientation of the target (105).

Description

"System and Method for Precise Real-Time Measurement of a Target Position and Orientation Relative to a Base Position, and Control Thereof

Field of the Invention

This invention relates to measurement of position in space of a target, and in particular to measurement of position and orientation in space of a target. This invention has particular application in measurement of position and orientation of a target relative to a base.

Background

The use of a laser tracker (LT) and a smart tracking system (STS) (together referred to as an LTSTS) to accurately determine the position and orientation of an object has been described in US patent 7,230,689 B2.

A laser tracker is an accurate measurement instrument that uses a laser distance measuring sensor and two accurate angle encoders to measure distance, bearing and azimuth to a target. A laser tracker takes a large number of measurements and uses an averaging algorithm to reduce the effect of noise to obtain an accurate measurement. The inventors have realised that this takes time and thus the accuracy increases as the measurement time increases. Where time is plentiful, a laser tracker will provide an accurate measure, but where time is constrained, the resulting loss of accuracy can be detrimental.

The smart tracking system is a target orientation sensor, and typically comprises a sensor that can "look back" at the laser tracker and derive pan and tilt data, and also roll data of the target.

Generally, the instantaneous data output from both the laser tracker and the smart tracking system can suffer from inaccuracy referred to as jitter, and needs to be time averaged, so that any jitter in the data is averaged out. LT and LTSTS systems are typically used for accurate measurements with accuracy between microns and mm and over relatively short ranges of typically less than 100m. Distance measurement is either by laser interferometry or time of flight measurement.

lnertial reference systems, also known as inertial navigation systems (INS) (comprising accelerometers and angular gyro) are used as navigation systems in aircraft for example. Inertial navigation systems suffer from drift and the accuracy from a starting reference point decreases with time. For this reason the relative position of an INS is updated with an absolute position from an external reference such as a GPS (Global Positioning System), in order to correct drift error.

Recursive filtering, such as a Kalman filtering algorithm can be used to reduce the noise in a data stream and also to combine two sets of data into one amalgamated data set. Such an arrangement is described in US patent application US2005/0060092, as applied to spacecraft rendezvous, combining INS measurements with measurements from a star tracker.

Normally inertial reference systems are used for large scale navigation and surveying tasks or for angular stabilisation tasks such as for pan and tilt cameras, radars and optical trackers. Honeywell International Inc.and iMAR GmbH supply such inertial reference systems. Typical inertial reference systems are used for tasks requiring accuracy similar to that provided by differential GPS, typically with errors greater than several mm but able to be used over ranges of up to thousands of km. Various types of accelerometers and gyros are used including ring laser gyros and micro-electro-mechanical systems (MEMS) accelerometers.

In addition, the use of multiple sensors in a redundant fashion to maintain data continuity in the event of data interuption from a sensor is also described in US Patent application US2005/0060092. However this does not enhance the accuracy of the data which is obtained. The preceding discussion of the background to the invention is intended to facilitate an understanding of aspects of the present invention. However, it should be appreciated that the discussion is not an acknowledgement or admission that any of the material referred to was part of the common general knowledge in Australia or elsewhere as at the priority date of the application.

It is an object of this invention to provide a system and method for precise realtime measurement of a target position and orientation relative to a base position which overcomes disadvantages inherent in the aforementioned configuration, or at least provides an alternative to the aforementioned arrangement.

Throughout the specification, unless the context requires otherwise, the word "comprise" or variations such as "comprises" or "comprising", will be understood to imply the inclusion of a stated integer or group of integers but not the exclusion of any other integer or group of integers.

Disclosure of the Invention

The inventors have found that combining measurements from a combined laser tracker and smart tracker system, and measurements from an inertial reference system, in a certain configuration, a fast and accurate measurement system can be created. The measurements of the LTSTS and INS can be combined so that the LTSTS compensates for the drift of the INS whilst the short term accuracy of the INS compensates for the noisy measurements of the LTSTS. The two signals can be combined, for example via a Kalman filter (or other computationally based combining algorithm) to give an accurate real time measurement.

The inventors have found that the resultant system is capable of measuring the position of an object with good accuracy at high speed, possibly of the order of microns and thousandths of a degree over ranges of up to 100m at an update rate of from 10 Hz up to 1 kHz, and ideally around 250 Hz to 300 Hz or thereabouts. It is expected that as technology advances the update rate will be able to be improved. The use of the method and apparatus of the invention allows the real time, non contact, accurate position measurement of industrial devices, and thus the closed loop position control of industrial devices such as construction machinery, measurement scanners, cameras, robots, vehicles and machine tools. This is achievable over a greater range and with greater accuracy than has previously been possible.

In accordance with one aspect of the invention there is provided an apparatus for precise real-time measurement of the position and orientation of a target relative to the position of a base, said base having a distance sensor with angle encoders arranged to send a beam to said target, to measure distance, altitude, and azimuth and output spatial position data of said target, said target having a target orientation sensor mounted on a pan and tilt mechanism, said target orientation sensor being arranged through said pan and tilt mechanism to reflect said beam from said distance sensor back to said distance sensor, said target orientation sensor being arranged to output orientation data being a measurement of orientation of said target, said apparatus also having an inertial reference system including motion sensors in or on said target to measure linear and angular acceleration in said target, said apparatus also including processor circuitry to derive first data being position and orientation data of said target from said distance sensor and angle encoders and from said target orientation sensor, with respect to time, said processor circuitry also separately processing measurements from said motion sensors to derive second data being position and orientation data of said target also with respect to time, said first data and said second data being combined by said processor circuitry to generate a single data set of position and orientation of said target with respect to time. In accordance with a second aspect of the invention there is provided an apparatus for precise real-time measurement and control of the position and orientation of a target relative to the position of a base, said base having a distance sensor with angle encoders arranged to send a beam to said target, to measure distance, altitude, and azimuth and output spatial position data of said target, said target having a target orientation sensor mounted on a pan and tilt mechanism, said target orientation sensor being arranged through said pan and tilt mechanism to reflect said beam from said distance sensor back to said distance sensor, said target orientation sensor being arranged to output orientation data being a measurement of orientation of said target, said apparatus also having an inertial reference system including motion sensors in or on said target to measure linear and angular acceleration in said target, said apparatus also including processor circuitry to derive first data being position and orientation data of said target from said distance sensor and angle encoders and from said target orientation sensor, also with respect to time, said processor circuitry also separately processing measurements from said motion sensors to derive second data being position and orientation data of said target with respect to time, said first data and said second data being combined by said processor circuitry to generate a single data set of position and orientation of said target with respect to time, said processor circuitry being used to control the position and orientation of said target in accordance with manually input or programmed data, and said data set being used as input in the control of the position and orientation of said target.

The distance, altitude and azimuth data may be a direct measure of distance, altitude and azimuth, or may be derived data providing the position of the target in three co-ordinates (x,y,z) from the base. The orientation data provides at least pitch and yaw data pertaining to the target, relative to the base. Preferably said first data and said second data are combined by said processor circuitry using a Kalman filter to generate said single data set of position and orientation of said target with respect to time.

Preferably said distance sensor measures distance with a laser interferometer.

Alternatively said distance sensor measures distance by time of flight measurement.

Preferably said target orientation sensor includes a roll sensor, to detect roll of said target, in addition to yaw and pitch (or pan and tilt).

Preferably said apparatus includes said base and said target as an integral unit.

Preferably the output data rate of time-averaged data output of the distance sensor with angle encoders lies in the range of from about 1 Hz to about 10 kHz.

Preferably the output data rate of time-averaged data output of the distance sensor with angle encoders lies in the range of from about 5 Hz to about 1 kHz.

Preferably the output data rate of time-averaged data output of the distance sensor with angle encoders lies in the range of from about 10 Hz to about 100 Hz.

Preferably the output data rate of time-averaged data output of the target orientation sensor encoders lies in the range of from about 1 Hz to about 10 kHz.

Preferably the output data rate of time-averaged data output of the target orientation sensor encoders is the same as the output data rate of time-averaged data output of the distance sensor with angle encoders.

Preferably the output data rate of the inertial reference system lies in the range of from 10 Hz to 100 kHz. Preferably the output data rate of the inertial reference system lies in the range of from 50 Hz to 10 kHz.

Preferably the output data rate of the inertial reference system lies in the range of from 100 Hz to 1 kHz.

Preferably the ratio of output data rate of time-averaged data output of the distance sensor with angle encoders and the target Orientation sensor, and the output data rate of the inertial reference system is from about 1 :5 to about 1 :100.

Preferably the ratio of output data rate of time-averaged data output of the distance sensor with angle encoders and the target orientation sensor, and the output data rate of the inertial reference system is from about 1 :10 to about 1 :50.

Preferably the ratio of output data rate of time-averaged data output of the distance sensor with angle encoders and the target orientation sensor encoders and the output data rate of the inertial reference system is about 1 :25.

It will be appreciated that the distance sensor with angle encoders, and the target orientation sensor, sample data internally at a data rate which is not necessarily related to the output data rate. Typically the rate at which successive measurements are taken or the internal sampling data rate of these instruments, will be very much higher that the output data rate. Successive measurements may be averaged over time for a whole or part of the period between successive output data.

The target which has the target orientation sensor and the inertial reference system may be a component, such as a robotic manipulator, a dolly for a camera, or a horizontally disposed platform, in which case the position and orientation of that component may be measured, and if required, controlled. Alternatively, the target may be the end of a boom or arm of a large machine, which can be moved through space, and whose orientation may change with change in altitude, and with extension in the case of an elbow-type arm mechanism. To the end of that boom or arm there may be fitted other equipment or componentry, such as a robotic manipulator, a dolly for a camera, or a horizontally disposed platform. In such an arrangement, further control systems can be provided to transform the position and orientation of the target to a desired position and orientation of the other equipment or componentry, by mapping the co-ordinates of the target to the desired co-ordinates of the other equipment or componentry.

Also in accordance with the second aspect of the invention there is provided a robotic arm mounted to a platform, said robotic arm having a coarse positioning robotic arm with an end which is controllably moveable towards and away from said platform and controllably adjustable in altitude and azimuth relative to said platform; said coarse positioning robotic arm being interfaced with an apparatus as hereinbefore described for precise real-time measurement and control of the position and orientation of a target relative to the position of a base, where the target is located at or near the end of the coarse positioning robotic arm, said end of said coarse positioning robotic arm having at least one fine positioning robotic arm adjustably mounted thereto on an end mount, said fine positioning robotic arm having a tool mount at an opposed position to said end mount; said processor circuitry also being used to control the position and orientation of said tool mount of said fine positioning robotic arm also in accordance with manually input or programmed data, and said position and orientation data set being used as input in the control of the position of said tool mount of said fine positioning robotic arm.

Preferably said base and said platform are discrete units, which in use are located a known distance from each other. Alternatively said base may be mounted on said platform. With this arrangement, deflection in the end of the coarse positioning robotic arm brought about for example by wind blowing against the coarse positioning robotic arm can be compensated by controlling the position and orientation of the tool mount of the fine positioning robotic arm.

Also in accordance with the first aspect of the invention there is provided a method for precise real-time measurement of the position and orientation of a target relative to the position of a base, the method comprising: providing said base with a distance sensor with angle encoders arranged to send a beam to said target, and to output distance, altitude, and azimuth data and measuring the distance, altitude, and azimuth from said base to said target, providing said target with a target orientation sensor mounted on a pan and tilt mechanism, controlling said target orientation sensor on said pan and tilt mechanism to reflect said beam from said distance sensor back to said distance sensor, and measuring pan and tilt angles to provide output orientation data being a measurement of orientation of said target, providing in said apparatus, an inertial reference system including motion sensors located in or on said target and measuring linear and angular acceleration in said target, deriving first data being position and orientation data of said target from said distance sensor and angle encoders and from said target orientation sensor, with respect to time, separately processing measurements from said motion sensors and deriving second data being position and orientation data of said target also with respect to time, combining said first data and said second data to generate a single data set of position and orientation of said target with respect to time.

Also in accordance with the second aspect of the invention there is provided a method for precise real-time measurement and control of the position and orientation of a target relative to the position of a base, the method comprising: providing said base with a distance sensor with angle encoders arranged to send a beam to said target, and to output distance, altitude, and azimuth data and measuring the distance, altitude, and azimuth from said base to said target, providing said target with a target orientation sensor mounted on a pan and tilt mechanism, controlling said target orientation sensor on said pan and tilt mechanism to reflect said beam from said distance sensor back to said distance sensor, and measuring pan and tilt angles to provide output orientation data being a measurement of orientation of said target, providing in said apparatus, an inertial reference system including motion sensors located in or on said target and measuring linear and angular acceleration in said target, deriving first data being position and orientation data of said target from said distance sensor and angle encoders and from said target orientation sensor, with respect to time, separately processing measurements from said motion sensors and deriving second data being position and orientation data of said target also with respect to time, combining said first data and said second data to generate a single data set of position and orientation of said target with respect to time, controlling the position and orientation of said target in accordance with manually input or programmed data, and using said data set as feed back in the controlling of the position and orientation of said target.

Preferably said first data is derived using processor circuitry. Processor circuitry may include, for example, any one of a micro-controller, microprocessor or dedicated hardware-based circuitry.

Preferably said processor circuitry also derives said second data. The processor circuitry may be an integral unit also deriving said first data, or may be a discrete unit including, for example, also any one of a micro-controller, microprocessor or dedicated hardware-based circuitry. Preferably said processor circuitry combines said first data and said second data to generate the single data set of position and orientation of said target with respect to time. Again the processor circuitry may be an integral unit also deriving said first data or said second data or both, or may be a discrete unit including, for example, also any one of a micro-controller, microprocessor or dedicated hardware-based circuitry.

The distance, altitude and azimuth data may be a direct measure of distance, altitude and azimuth, or may be derived data providing the position of the target in three co-ordinates from the base. The orientation data provides at least pitch and yaw data pertaining to the target, relative to the base.

Preferably said first data and said second data are combined by said processor circuitry using a Kalman filter to generate said single data set of position and orientation of said target with respect to time.

Preferably said distance sensor measures distance with a laser interferometer.

Preferably the output data rate of time-averaged data output of the distance sensor with angle encoders lies in the range of from about 5 Hz to about 1 kHz. Preferably the output data rate of time-averaged data output of the distance sensor with angle encoders lies in the range of from about 10 Hz to about 100 Hz.

Preferably the output data rate of time-averaged data output of the target orientation sensor lies in the range of from about 1 Hz to about 10 kHz.

Preferably the output data rate of time-averaged data output of the target orientation sensor is the same as the output data rate of time-averaged data output of the distance sensor with angle encoders.

Preferably the output data rate of the inertial reference system lies in the range of from 10 Hz to 100 kHz.

Preferably the output data rate of the inertial reference system lies in the range of from 50 Hz to 10 kHz.

Preferably the ratio of output data rate of time-averaged data output of the distance sensor with angle encoders and the target orientation sensor encoders and the output data rate of the inertial reference system is about 1 :25. Brief Description of the Drawings

Four preferred embodiments of the invention will now be described in the following description of apparatus for measuring the position and orientation of a target relative to a base, made with reference to the drawings in which:

Figure 1 is a schematic diagram of the control circuitry architecture for the first embodiment;

Figure 2 is a schematic diagram of the control circuitry architecture for the second embodiment;

Figure 3 is a schematic diagram of the control circuitry architecture for the third embodiment; Figure 4 is a schematic diagram of the control circuitry architecture for the fourth embodiment of the invention;

Figure 5 is an orthographic diagram showing the apparatus of the fourth embodiment in an application for controlling the position and orientation of the end of a boom; Figure 6 is an orthographic view showing the apparatus of the fourth embodiment in use on a boom for a robotic arm; and

Figure 7 is a part orthographic view of part of the robotic arm shown in figure 6, illustrating an alternative embodiment.

Best Mode(s) for Carrying Out the Invention

All of the embodiments are apparatus for precisely measuring in real-time, the position and orientation of a target relative to a base. The resultant position and orientation data may then be put to use, by being fed to a display for showing the position and orientation data for the target, and/or put to use as feedback in a control system for controlling the position and orientation of the target or a device physically connected to the target.

In the following descriptions of the first three embodiments, which are implementations of the invention in slightly different ways, some off-the-shelf componentry is used. Specific model numbers of manufacturers off-the-shelf equipment is used in the description to provide examples of physical implementation. From this information, persons skilled in the art will appreciate that alternative equipment could be used and would be within the scope of the present invention.

In a first embodiment an Automated Precision Inc. "OmniTrac™" time of flight laser tracker being a distance sensor with angle encoders is used in conjunction with an Automated Precision Inc. "SmartTRACK™" orientation sensor, together indicated at 11 in figure 1 , to provide first data. An IMAR iNAV-RQH-1004 is used as an inertial reference system 13 to provide second data. The laser tracker is mounted on a base, and the orientation sensor is mounted on the target. The inertial reference system is also mounted on the target.

The laser tracker is an accurate measurement instrument that uses a laser distance measuring sensor and two accurate angle encoders to measure distance, altitude and azimuth to a target. The laser tracker sends a beam of polarised light from the base to the target. The orientation sensor provides an accurate measurement of the orientation of a target, and effectively comprises a sensor that is mounted on a pan and tilt mechanism that "looks back" at the laser tracker, reflecting the beam of light back to the laser tracker and measurements are made of pan and tilt angles at the orientation sensor. The orientation sensor can also make a roll measurement through use of a roll sensor using a polarising filter which is rotated, and measurement of the rotation is taken to determine the angle of polarisation of the light from the laser tracker, and hence the rotational angle of the target relative to the base.

In an alternative embodiment, the distance sensor used on a laser tracker may be a laser interferometer such as an Automated Precision Inc. "Tracker3™" laser tracker. A time of flight distance sensor as used in the embodiments described here, transmits a pulse of laser light that is reflected off a target and returns to the sensor. The time of flight is measured and knowing the speed of light allows the distance to be calculated. A laser interferometer laser tracker sends a beam of light through a beam splitter prism that sends a first beam to a target mirror that retro reflects the light of the first beam back to the interferometer. The first beam is then combined against the second beam from the beam splitter prism and the resultant beam experiences an interference pattern due to the phase difference of the coherent light in the two beams. As the target mirror is moved towards or away from the interferometer the phase relationship of the two beams changes. The interferometer has a sensor that measures the interference pattern of the two beams and counts the "interference fringes". With a laser interferometer laser tracker, the measurement must start with the target mirror located at a precise known distance relative to the interferometer and then distance changes are measured by counting the "interference fringes". In such an alternative embodiment the laser light first passes thorough a polarising filter so that the light beam is polarised and the angle of polarisation can be detected by the roll sensor of the orientation sensor in the target.

As stated above, the orientation sensor looks back at the laser tracker light beam. Light enters the orientation sensor telescope and passes through a beam splitting prism. The first beam passes through a second beam splitter prism to create a third beam. The first beam then travels to a retro reflective mirror that provides the target for the distance sensor and reflects the first beam back to the distance sensor. The second beam passes to a CCD (charge coupled device) sensor array that is used as a control signal to the controller that moves the pan and tilt motors to make the orientation sensor track the laser tracker. The third beam passes to the roll sensor. The roll sensor determines the polarisation angle of the laser light and therefore provides a roll measurement for the orientation sensor.

Both the combined laser tracker and the orientation sensor process their data via a single internal computer. The position and orientation data is time averaged, and output 17 at a data rate of 10 Hz. The inertial reference system also has an internal computer for processing its position and orientation data, which is output 19 at a data rate of 250 Hz. The combined laser tracker and the orientation sensor and inertial reference system communicate digitally on an ethemet network 21 with a combining computer 23, which combines the data. The data from the combined laser tracker and the orientation sensor is used to correct for gyroscopic drift inherent in the inertial reference system.

The measurement data from the combined laser tracker and the orientation sensor 11 is time stamped. The measurement data from the inertial reference system 13 is also time stamped. The combining computer ensures that the time stamps are synchronised via the computer clocks. The computer clocks are synchronised by a TTL logic output 25 provided by the synchronising computer

23 to both the inertial reference system 13 and the combined laser tracker and the orientation sensor 11.

The measurement data from the combined laser tracker and the orientation sensor, and the inertial reference system is combined via a Kalman filter (or other suitable combination algorithm) to generate a single set of time stamped position, velocity and acceleration data for all six degrees of freedom. The Kalman filter must be tuned to provide optimum output. To understand the filtering process in lay terms, an approximation is that the combined laser tracker and the orientation sensor data is low pass filtered while the inertial reference system data is high pass filtered.

By way of explanation, the optimum Kalman filter combines a predicted data set with a measured data set to give a filtered data set. The predicted data set is obtained by using the previously obtained filtered data set and a model of the system combined with control data and structural dynamics information to predict the current data set. The measured data set is obtained from the instrumentation. The combination algorithm first compares the measured data with the predicted data to check that the measured data is within range and does not contain excessive noise or null values. If the measurements are out of range they are trimmed back to the limit value. If the measurements are null values they are set to the predicted value. The next step of the algorithm calculates a weighted average of the predicted data and the measured data. For the optimum Kalman Filter the covariance matrix of the measurement variables can be used to determine the optimum Kalman Filter gain. Alternatively optimised values can be determined by experiment.

In the six degree of freedom case the system is governed by Newtons laws of motion. In the general case the mass and intertia may change with time due to changes in machine (target) pose, configuration or payload. Measurements from the inertial reference system are in units of acceleration and therefore the predicted motion is independent of mass and inertia changes. The control inputs are generally in terms of force input (via actuators) and therefore the resultant motion does depend on the machine mass and inertia. Structural dynamics are pose, configuration, mass and inertia dependant. The structural dynamics can be considered as the sum of static deflection and oscillatory response. Whilst it is possible to model all of these effects it is computationally efficient to make simplifications that have little effect on the accuracy.

In an ideal application of this technology, the mass is large compared to the exciting force so that the acceleration due to control or disturbance will be low.

A linear forward prediction over a short timeframe, of eg 4ms, experiences effectively no influence from any external or internally applied force because acceleration has so little effect on displacement (s=ut+0.5at²).

Peak accel = 4ms^"2, t=0.004s, displacement = s = 0.032mm. The relative displacement from control cycle to cycle really depends only on the current velocity. The displacement due to acceleration is a second order effect and can be discounted. This means that short term latency or delays in the control cycles are not a problem provided they are consistent. At 10ms the potential error may start to become significant. Velocity and latency has a much greater effect, for example at 600mm/sec, a 1 ms variation in latency would change the predicted position by 0.6mm (ie 2Ox the effect of the worst case external transient force). This is advantageous because it means that all of the uncertainty due to acceleration variation is of second order effect and of little consequence over one time cycle. The system state model is thereby reduced to one of position and velocity while the measurements are of position by the combined laser tracker and orientation sensor and acceleration measured by the inertial reference system. The inertial reference system acceleration measurements are used for the prediction.

Regarding the level of accuracy achieved, the combined laser tracker and orientation sensor achieves 10um ± 1 ppm and outputs every 10ms. The inertial reference system can output every 1 ms and a drift rate of 1000mm/hr which over 10ms has drift of the order of 2.7μm. Therefore the combined system can output measurements at 1 ms interval with an error not exceeding 13μm ± 1 ppm.

Referring to figure 2, in a second embodiment an Automated Precision Inc. "OmniTrac™" laser tracker being a distance sensor with angle encoders is used in conjunction with an Automated Precision Inc. "SmartTRACK™" orientation sensor as the combined laser tracker and orientation sensor indicated at 11. An IMAR iNAV-RQH-1004 is used as the inertial reference system 13.

As in the first embodiment, the laser tracker is mounted on a base, and the orientation sensor is mounted on the target. The inertial reference system is also mounted on the target.

Both the combined laser tracker and the orientation sensor and inertial reference system have their own internal computers. The combined laser tracker and the orientation sensor and inertial reference system communicate digitally on an ethernet network 21 with each other.

The measurement data, averaged and output at a data rate of 10 Hz by the combined laser tracker and the orientation sensor, is time stamped. The inertial reference system 13 computer acts as a master and synchronises the combined laser tracker and the orientation sensor clock with its own clock. Depending on the communication protocol used, a separate hardware signal may be required to synchronise the clocks. This hardware signal may be a TTL output 25 from the master that is read as an input by the other computer. The inertial reference system outputs data at a rate of 250 Hz. The inertial reference system computer receives data from the combined laser tracker and the orientation sensor computer and then performs the calculations to combine the data.

The measurement data from the combined laser tracker and the orientation sensor and the inertial reference system is combined via a Kalman filter (or other suitable combination algorithm) to generate a single set of timestamped position, velocity and acceleration data for all six degrees of freedom. The Kalman filter must be tuned to provide optimum output.

Referring to figure 3, in a third embodiment an Automated Precision Inc. "OmniTrac™" laser tracker being a distance sensor with angle encoders is used in conjunction with an Automated Precision Inc. "SmartTRACK™" orientation sensor as the combined laser tracker and the orientation sensor 11. An IMAR iNAV-RQH-1004 is used as the inertial reference system 13. As in the first embodiment, the laser tracker is mounted on a base, and the orientation sensor is mounted on the target. The inertial reference system is also mounted on the target.

The measurement data at an output data rate of 250 Hz from the inertial reference system is time stamped. The combined laser tracker and the orientation sensor averages its measured data and provides output data at a data rate of 10 Hz. The combined laser tracker and the orientation sensor computer acts as a master and synchronises the inertial reference system clock with its own clock. Depending on the communication protocol used a separate hardware signal may be required to synchronise the clocks. This hardware signal may be a TTL output 25 from the master that is read as an input by the other computer. The combined laser tracker and the orientation sensor computer receives data from the INS computer and then performs the calculations to combine the data.

In the first three embodiments, the communication interface between the computers must be time synchronised. In these embodiments the various computers communicate by a real time deterministic network (such as Ethercat, Sercos, Sercosll or Sercoslll, Profibus, DeviceNet, Powerlink or Synqnet). Alternatively one computer may use a high speed hardware output that is read as a high speed input by the other computer. This hardware signal is switched at a known time, therefore allowing synchronisation of the two computer clocks.

Referring to figures 4 and 5, in a fourth and most preferred embodiment the basic sensors and control systems that are normally employed separately in an inertial reference system and a combined laser tracker and orientation sensor, are combined and controlled by a single computer 100. Figure 4 shows in simplified form the logical connection between the sensors and the computer 100 which performs the combination algorithm. Raw data values are buffered so that the data is immediately available to the computer when required. Digital filtering used individually on each channel to filter and condition the digital data. The input of data to the computer and the initial filtering of individual inputs utilises standard known techniques.

Referring to figure 5, a laser tracker 101 provided by an Automated Precision Inc. "OmniTrac™" time of flight distance sensor with angle encoders is mounted to a base in the form of a tripod 102, and an orientation sensor 103 provided by an Automated Precision Inc. "SmartTRACK™" orientation sensor, and an inertial reference system 104 being an IMAR iNAV-RQH-1004 are mounted to a target or head 105.

The laser tracker 101 includes a pan and tilt angle measurement mechanism 106 and a range finder 107. The laser tracker 101 has a first pan axis 109 and a second tilt axis 111. The laser tracker 101 has a time of flight range-finder 107, and sends a beam of polarised light to a target mirror (contained in the orientation sensor 103) that retro reflects the light of the first beam back to the rangefinder.

The orientation sensor 103 includes a pan 113, tilt 115 and roll tracking mechanism 117 that looks back at the laser tracker 101. The orientation sensor 103 has a first pan axis 119, a second tilt axis 121 and a third roll axis 123. The head 105 also includes three orthogonal angular gyroscopes 125, 127 and 129 and three orthogonal linear accelerometers 131 , 133 and 135. The roll tracking mechanism 117 detects the polarised light emitted by the polarised light source of the laser tracker 101 and provides a roll angle relative to the base laser tracker 101 (or horizontal plane through the second tilt surface - assuming the tripod 102 is mounted horizontally on horizontal ground). In the preferred embodiment the angular gyroscopes 125, 127 and 129 are ring laser gyroscopes but in other embodiments could be any other type of suitable angular or rate gyroscope such as a strapped down MEMS gyroscope, galvanometer or mechanical (spinning disk) gyroscope, or fibre optic gyroscope.

All rotary axes on the pan and tilt mechanisms 106 and 113, 115 have high accuracy encoders (not shown) that provide accurate digital angle measurement and servo motors (also not shown) that provide accurate motion. The servo motors are connected to amplifiers 137 via cables 139 and 141 so that the tracking system operates in a closed loop according to well known industrial motion control methods. In a most preferred arrangement, the servo motors are direct drive torque motors. The encoders provide angle measurement to the computer 100. The computer 100 combines the angle measurements received from the axes 109, 111 , 119, 121 and 123 and the distance measurement from the range finder 107 to determine the relative position and orientation of the head 105 from the base 102. The position of the base 102 relative to a world datum can be determined by moving the head to that datum and back calculating. At least three world datums are required to provide all six positional coordinates. This is done by known surveying principles and trigonometry. By combining the known base 102 position with the relative head 105 position, the absolute direct position of the head 105 can be calculated.

The computer 100 also receives acceleration data from the linear accelerometers 131 , 133 and 135 and orientation or orientation acceleration (depending on the type of sensor 125-129) information from the angular gyros 125, 127 and 129. By using a dead reckoning algorithm the computer 100 integrates the acceleration data over time including orientation effects to obtain velocity information and integrates the velocity information over time including orientation effects to obtain inertial relative positional information.

The computer 100 then repetitively combines the inertial relative positional information with the absolute direct position through a Kalman or other suitable filtering algorithm to obtain a combined corrected absolute position and orientation.

Referring to Figure 6 a non contact first measurement instrument in the form of the laser tracker 101 set on the tripod 102 is set up so that it is fixed relative to a workpiece 202. In this preferred embodiment the workpiece 202 is rested on the ground 203 and the laser tracker 101 set on the tripod 102 rested on the ground.

A robotic arm 204 is set on a platform 205 which may or may not be directly attached to the workpiece 202. In the preferred embodiment the platform 205 rests on the ground 203. The base 205 supports a coarse positioning robotic arm comprising a telescopic boom 206 with a vertically travelling column 208 at the end 207 of the boom 206.

The telescopic boom 206 is mounted to the platform 205 about a joint which may pivot about both a horizontal axis allowing the boom to be raised and lowered (altitude), and a vertical axis allowing the boom to be rotated horizontally (azimuth).

Those skilled in the art will appreciate that the coarse positioning robotic arm could consist of any type of mechanism that can move a second end relative to a first end. Such mechanisms include but are not limited to industrial robots, cranes, booms, telescopic booms, SCARA arms, overhead gantry or manipulators with any combination of articulated and sliding joints, gantries or machine tools.

The target or head 105 is attached at the end of the telescopic boom 206 to the vertically travelling column 208, and so measures the combined position and orientation of the coarse positioning robotic arm 206 and vertically travelling column 208.

The vertically travelling column 208 has mounted there to, base 209 to which is attached a fine positioning robotic arm 210. The fine positioning robotic arm 210 has a horizontal bar 211 which moves slidingly in a horizontal manner relative to the base 209 and vertically travelling column 208, and also moves rotatably about the base 209. The fine positioning robotic arm 210 includes at an end located away from the base 209 connection with the vertically travelling column 208, a robot manipulator 212 having a tool mount 213 in which is mounted a robotic gripper 214.

The fine positioning robotic arm 210 may be any type of robot or manipulator that allows movement in at least five and preferably six degrees of freedom. In the embodiment shown the fine positioning robotic arm 210 is an r, theta, z robot manipulator with a three axis wrist.

In an alternative embodiment, the fine positioning robotic arm may incorporate the vertically travelling column 208, and the target or head 105 can be attached directly to the end of the telescoping boom 206.

The laser tracker 101 and the target or head 105 communicatively connected 141 , 139 to the computer 100 which includes a measurement processing unit 215. The communication connection may be by physical connection such as a serial data cable as shown in figure 2, or ethernet cable or fibre optic or by a data transmission wireless link, in alternative arrangements .

The measurement system processing unit 215 calculates the position and orientation in real time of the vertically travelling column 208 to which the fine positioning robotic arm 210, 212 is attached.

The coarse positioning robotic arm formed by the telescopic boom 206 and the vertically travelling column 208 is controlled by a control computer system 217. The measurement system processing unit 215 communicates the position and orientation of the vertically travelling column 208 to the control computer system 217. The control computer system 217 compares the actual position and orientation of the vertically travelling column 208 to the desired position and orientation and thereby calculates a six degree of freedom error vector. The control computer system 217 then calculates the required axis positions and motion parameters of the fine positioning robotic arm comprising the fine positioning robotic arm 210 and the robot manipulator 212, taking into account the six degree of freedom error vector so that the tool mount 213 and robotic gripper 214 are positioned at the required orientation relative to the workpiece 202. Acceleration measurements available from linear accelerometers and rate gyros, are used to calculate acceleration feedforward which is combined using the Kalman filter previously discussed, to improve motion dynamics. The axis positions and motion parameters are then used by motion control 219 to move the axes to the required positions. The motion control 219 is connected via cables 221 and hoses 223 to the telescopic boom 206 and the vertically travelling column 208 and fine positioning robotic arm 210, the robot manipulator 212 and the gripper 214.

Referring to figure 3, an alternative vertically travelling column 208 and fine positioning robotic arm 210 is shown. The base 209 is mounted on a track 231 located in the vertically travelling column 208 for fine vertical positioning of the fine positioning robotic arm 210. The base 209 is connected to the horizontal bar 211 of the fine positioning robotic arm 210 by a column 233 which stands off the horizontal bar 211 from the vertically travelling column in order to provide clearance when the base 209 is raised up the track 231. A carriage 235 is rotatably fitted to the bottom of the column 233, the carriage providing sliding support for the horizontal bar 211.

Those skilled in the appropriate arts will appreciate that multiple fine positioning robotic arms could be incorporated so that multiple tasks may be undertaken. Various delivery mechanisms such as pipes and hoses or conveyor belts may be added to the boom 206 to deliver materials and or components or tooling to the end 207 of the boom 206 or the column 208 so that they may be used by the gripper 214.

Those skilled in the art will appreciate that numerous other measurement and correction methods may be employed and combined without deviating from the fundamental inventive step which is to combine 6 degree of freedom inertial measurements with six degree of freedom direct optical angle and distance measurements to yield positional data of greater accuracy and update rate than can be gained by employing either method in isolation.

It should be appreciated that the scope of the invention is not limited to the particular embodiment disclosed herein.

Claims

The Claims Defining the Invention are as Follows

1. An apparatus for precise real-time measurement of the position and orientation of a target relative to the position of a base, said base having a distance sensor with angle encoders arranged to send a beam to said target, to measure distance, altitude, and azimuth and output spatial position data of said target, said target having a target orientation sensor mounted on a pan and tilt mechanism, said target orientation sensor being arranged through said pan and tilt mechanism to reflect said beam from said distance sensor back to said distance sensor, said target orientation sensor being arranged to output orientation data being a measurement of orientation of said target, said apparatus also having an inertial reference system including motion sensors in or on said target to measure linear and angular acceleration in said target, said apparatus also including processor circuitry to derive first data being position and orientation data of said target from said distance sensor and angle encoders and from said target orientation sensor, with respect to time, said processor circuitry also separately processing measurements from said motion sensors to derive second data being position and orientation data of said target also with respect to time, said first data and said second data being combined by said processor circuitry to generate a single data set of position and orientation of said target with respect to time.

2. An apparatus for precise real-time measurement and control of the position and orientation of a target relative to the position of a base, said base having a distance sensor with angle encoders arranged to send a beam to said target, to measure distance, altitude, and azimuth and output spatial position data of said target, said target having a target orientation sensor mounted on a pan and tilt mechanism, said target orientation sensor being arranged through said pan and tilt mechanism to reflect said beam from said distance sensor back to said distance sensor, said target orientation sensor being arranged to output orientation data being a measurement of orientation of said target, said apparatus also having an inertial reference system including motion sensors in or on said target to measure linear and angular acceleration in said target, said apparatus also including processor circuitry to derive first data being position and orientation data of said target from said distance sensor and angle encoders and from said target orientation sensor, also with respect to time, said processor circuitry also separately processing measurements from said motion sensors to derive second data being position and orientation data of said target with respect to time, said first data and said second data being combined by said processor circuitry to generate a single data set of position and orientation of said target with respect to time, said processor circuitry being used to control the position and orientation of said target in accordance with manually input or programmed data, and said data set being used as input in the control of the position and orientation of said target.

3. An apparatus as claimed in claim 1 or 2 wherein said first data and said second data are combined by said processor circuitry using a Kalman filter to generate said single data set of position and orientation of said target with respect to time.

4. An apparatus as claimed in any one of the preceding claims wherein said target orientation sensor includes a roll sensor, to detect roll of said target, in addition to yaw and pitch (or pan and tilt).

5. An apparatus as claimed in any one of the preceding claims wherein the output data rate of time-averaged data output of the distance sensor with angle encoders lies in the range of from about 1 Hz to about 10 kHz.

6. An apparatus as claimed in claim 5 wherein the output data rate of time- averaged data output of the target orientation sensor encoders is the same as the output data rate of time-averaged data output of the distance sensor with angle encoders.

7. An apparatus as claimed in any one of the preceding claims wherein the output data rate of the inertial reference system lies in the range of from 10 Hz to 100 kHz.

8. An apparatus as claimed in any one of the preceding claims wherein the ratio of output data rate of time-averaged data output of the distance sensor with angle encoders and the target orientation sensor, and the output data rate of the inertial reference system is from about 1 :5 to about 1 :100.

9. An apparatus as claimed in claim 8 wherein the ratio of output data rate of time-averaged data output of the distance sensor with angle encoders and the target orientation sensor encoders and the output data rate of the inertial reference system is about 1 :25.

10. A robotic arm mounted to a platform, said robotic arm having a coarse positioning robotic arm with an end which is controllably moveable towards and away from said platform and controllably adjustable in altitude and azimuth relative to said platform; said coarse positioning robotic arm being interfaced with an apparatus as claimed in any one of claims 2 to 9 for precise real-time measurement and control of the position and orientation of a target relative to the position of a base, where the target is located at or near the end of the coarse positioning robotic arm, said end of said coarse positioning robotic arm having at least one fine positioning robotic arm adjustably mounted thereto on an end mount, said fine positioning robotic arm having a tool mount at an opposed position to said end mount; said processor circuitry also being used to control the position and orientation of said tool mount of said fine positioning robotic arm also in accordance with manually input or programmed data, and said position and orientation data set being used as input in the control of the position of said tool mount of said fine positioning robotic arm.

11. A method for precise real-time measurement of the position and orientation of a target relative to the position of a base, the method comprising: providing said base with a distance sensor with angle encoders arranged to send a beam to said target, and to output distance, altitude, and azimuth data and measuring the distance, altitude, and azimuth from said base to said target, providing said target with a target orientation sensor mounted on a pan and tilt mechanism, controlling said target orientation sensor on said pan and tilt mechanism to reflect said beam from said distance sensor back to said distance sensor, and measuring pan and tilt angles to provide output orientation data being a measurement of orientation of said target, providing in said apparatus, an inertial reference system including motion sensors located in or on said target and measuring linear and angular acceleration in said target, deriving first data being position and orientation data of said target from said distance sensor and angle encoders and from said target orientation sensor, with respect to time, separately processing measurements from said motion sensors and deriving second data being position and orientation data of said target also with respect to time, combining said first data and said second data to generate a single data set of position and orientation of said target with respect to time.

12. A method for precise real-time measurement and control of the position and orientation of a target relative to the position of a base, the method comprising: providing said base with a distance sensor with angle encoders arranged to send a beam to said target, and to output distance, altitude, and azimuth data and measuring the distance, altitude, and azimuth from said base to said target, providing said target with a target orientation sensor mounted on a pan and tilt mechanism, controlling said target orientation sensor on said pan and tilt mechanism to reflect said beam from said distance sensor back to said distance sensor, and measuring pan and tilt angles to provide output orientation data being a measurement of orientation of said target, providing in said apparatus, an inertial reference system including motion sensors located in or on said target and measuring linear and angular acceleration in said target, deriving first data being position and orientation data of said target from said distance sensor and angle encoders and from said target orientation sensor, with respect to time, separately processing measurements from said motion sensors and deriving second data being position and orientation data of said target also with respect to time, combining said first data and said second data to generate a single data set of position and orientation of said target with respect to time, controlling the position and orientation of said target in accordance with manually input or programmed data, and using said data set as feed back in the controlling of the position and orientation of said target.

13. A method as claimed in claim 1 1 or 12 wherein said first data is derived using processor circuitry.

14. A method as claimed in claim 13 wherein said processor circuitry also derives said second data.

15. A method as claimed in claim 14 wherein said processor circuitry combines said first data and said second data to generate the single data set of position and orientation of said target with respect to time. .

16. A method as claimed in claim 15 wherein said first data and said second data are combined by said processor circuitry using a Kalman filter to generate said single data set of position and orientation of said target with respect to time.

17. A method as claimed in any one of claims 11 to 16 wherein said target orientation sensor includes a roll sensor, to detect roll of said target, in addition to yaw and pitch (or pan and tilt).

18. A method as claimed in any one of claims 11 to 16 wherein the output data rate of time-averaged data output of the distance sensor with angle encoders lies in the range of from about 1 Hz to about 10 kHz.

19. A method as claimed in claim 18 wherein the output data rate of time- averaged data output of the target orientation sensor is the same as the output data rate of time-averaged data output of the distance sensor with angle encoders.

20. A method as claimed in any one of claims 11 to 19 wherein the output data rate of the inertial reference system lies in the range of from 10 Hz to 100 kHz.

21. A method as claimed in claim 20 wherein the ratio of output data rate of time-averaged data output of the distance sensor with angle encoders and the target orientation sensor, and the output data rate of the inertial reference system is from about 1 :5 to about 1 :100.

22. A method as claimed in claims 21 wherein the ratio of output data rate of time-averaged data output of the distance sensor with angle encoders and the target orientation sensor encoders and the output data rate of the inertial reference system is about 1 :25.

23. An apparatus for precise real-time measurement of the position and orientation of a target relative to the position of a base, substantially as herein described with reference to the description of the embodiment.

24. An apparatus for precise real-time measurement and control of the position and orientation of a target relative to the position of a base, substantially as herein described with reference to the description of the embodiment.